Grok 4.6 Ports Python Library to Rust Following Fable Plan
DHH reports using the model to replicate a code rewrite task at reduced expense.
DHH posted that Grok 4.6 followed Fable's plan to rewrite a Python library in Rust. With a couple of nudges the model completed the work. DHH observes that the effort used Fable's plan rather than starting from scratch. He calls the outcome encouraging. Gavin Baker responded by emphasizing that the future likely involves multiple models working together and that lower margins at the model layer will increase demand for AI infrastructure.
I had @spacexai Grok 4.6 follow Fable's plan, and with just a couple of nudges, it was able to repeat this feat in 1h 24m using 8.6M tokens at a cost of ~$55 at per-token pricing. That's about 1/10 the cost of the Fable implementation for the same work!
Fable one-shotted a Rust rewrite of the TerminalTextEffects Python library in 11M tokens. Startup time went from 87ms to 2ms and rendering speed is up by 9.6x. Now zero dependencies and a 3mb single exec 🤯
Reactions from ranked influencers
3 posts@SpaceXAI Again, to be fair, this was Fable's plan. I haven't tried asking Grok to do it entirely from scratch, including planning. But if someone wants to spend the tokens, I'll republish the results! Either way, super encouraging that Grok 4.6 is able to do this 🤘🤘
Important. The future is almost certain to be multiple models working together. Lower margins at the model layer whether from open-source or vertically integrated players = more AI infrastructure demand.
I had @spacexai Grok 4.6 follow Fable's plan, and with just a couple of nudges, it was able to repeat this feat in 1h 24m using 8.6M tokens at a cost of ~$55 at per-token pricing. That's about 1/10 the cost of the Fable implementation for the same work!
Fable one-shotted a Rust rewrite of the TerminalTextEffects Python library in 11M tokens. Startup time went from 87ms to 2ms and rendering speed is up by 9.6x. Now zero dependencies and a 3mb single exec 🤯